dmi modeling systems and plans for ceeh activities off-line air pollution modeling on-line air...
Post on 21-Dec-2015
221 views
TRANSCRIPT
DMI Modeling Systems And Plans For CEEH
Activities
•Off-Line Air Pollution Modeling•On-Line Air Pollution Modeling•Emergency Preparednes &
Risk Assessment•Urban Modeling
EnergyEnergy EnvironmentEnvironment HealthHealthCC Kick-off.
d. 23-25/1 2007
A. Gross, A. Baklanov, U. S. Korsholm, J. H. Sørensen, A. Mahura & A. Rasmussen
Content:
Air Pollution Modeling At DMI EnergyEnergy EnvironmentEnvironment HealthHealthCC
1. PSC aerosols2. Tropospheric
aerosols
Approaches:Normal distribution,Bin approach
Physics:1. Condensation2. Evaporation3. Emission4. Nucleation5. Deposition
Aerosol Module1. Gas Phase2. Aqueous phase3. Chemical equil.4. Climate Modeling
Approaches:RACM, CBIV, ISORROPIA
Chemical Solvers
Lagrangiantransport, 3-Dregional scale
UTLS Trans. Models
Eulerian trans-port 0..15lat-lon grid,3-D regional scale
ECMWF
DMI-HIRLAM
Eulerian trans-port 0.2-0.05lat-lon, 25-40 vert. layer, 3-D regional scale
StochasticLagrangian transport,3-D regional scale
On-Line Chemical Aerosol Trans.
ENVIRO-HIRLAM
Off-Line Chemical Aerosol Trans.
CAC
Emergency Pre-parednes & Risk Assess-
ment. DERMA
Nuclear, veterinary and chemical.
Regional (European) to city scale air pollution: smog and ozone.
Regional (European) scale air pollution: smog and ozone, pollen.
Tropo. Trans. Models
Met. Models
City-Scale Obstacle Resolved Modelling
TSU-CORM
DMI-HIRLAM
A forecast integration starts out by assimilation of meteorological observations whereby a 3-d state of the atmosphere is produced, which as well as possible is in accordance with the observations.
Currently nested versions of HIRLAM:• T – 15x15 km2, 40 vertical layers.• S – 5x5 km2, 40 vertical layers. • Q – 5x5 km2, 40 vertical layers.• Test version of 1.5x1.5 km2 of DK.
A numerical weather prediction system consists of pre-processing, climate file generation, data-assimilation and analysis, initialization, forecast, post-processing and verification.
TS
Q
EnergyEnergy EnvironmentEnvironment HealthHealthCC
Climate Change Scanarios Modeling By HIRHAM
EnergyEnergy EnvironmentEnvironment HealthHealthCC
Modeling Area
Simulation period: Year 2000 to 2100
Output of meteorological parameter:
From 3-6 hours to once a day depend
on the parameter.
•Horizontal resolution 25x25 km2.
•Vertical resolution 19 levels.
Off-Line modelling with CAC
Simulation domain
Horizontal resolution 0.2º×0.2º.
T:0.15º×0.15º
S: 0.05º×0.05º
EnergyEnergy EnvironmentEnvironment HealthHealthCC
CAC Model Area
ENSEMBLE JRC project exp. nr. 11 EnergyEnergy EnvironmentEnvironment HealthHealthCC
(Off-Line)
Ensemble: DK3, DE1, FR2, CA2
Ozone36 hour forecast 48 hour forecast
0 15 30 60 90 120 150
ppbV
EnergyEnergy EnvironmentEnvironment HealthHealthCC
(Off-Line)
“Semi”-operational forecasts 4 times a day of O3, NO, NO2, CO, SO2, Rn, Pb, “PM2.5”, “PM10”.
Advantages of On-line & Off-line modeling
On-line coupling• Only one grid; No interpolation
in space• No time interpolation• Physical parameterizations are
the same; No inconsistencies• Possibility of feedbacks
bewte-en air pollution and meteoro-logy
• All 3D met. variables are ava-ilable at the right time (each time step); No restriction in variability of met. fields
• Does not need meteo- pre/postpro-cessors
Off-line• Possibility of independent
parame-terizations• Low computational cost; • More suitable for ensembles
and oprational activities • Independence of atmospheric
pol-lution model runs on meteorolo-gical model computations
• More flexible grid construction and generation for ACT models
EnergyEnergy EnvironmentEnvironment HealthHealthCC
Radiation budgets
Temperature profiles
Chemistry/Aerosols
CloudCondensation
Nuclei
Precipitation
Chemistry/Aerosols
Examples of feedbacks
Cloud-radiationinteraction
Temperature profiles
Chemistry/Aerosols
EnergyEnergy EnvironmentEnvironment HealthHealthCC
Emission
Transport
Dispersion
Deposition
Gas phase chemistry
Aerosol chemistry
Aerosol physics
CloudsPrecipitation
RadiationDMI-HIRLAM
U, V, W, T, q,
U*, L
Concentration/Mixing ratio
On-Line Modeling With ENVIRO-HIRLAM
EnergyEnergy EnvironmentEnvironment HealthHealthCC
EnergyEnergy EnvironmentEnvironment HealthHealthCC Chernobyl Simulation 0.15°x0.15°, d. 7/5-1986, 18.00 UTC
Dry deposition (kBq/m2)
Total deposition statistics: Corr = 0.59, NMSE 6.3
Accumulated (reference) dry deposition [μg/m2] +48 h Difference (ref – perturbation) inAccumulated dry deposition [ng/m2]
Emergency Preparednes & Risk Assessment
Using the 3-D Stochastic Lagragian Regional Scale Model DERMA
EnergyEnergy EnvironmentEnvironment HealthHealthCC
1. Probabilistic Risk Assessment.
2. Source Determination by Inverse Modelling.
3. Chemical Emergency Preparednes.
4. Urban Meteorology Effects.
Examples:
Probabilistic Risk Assessment
Yearly time-integrated concentration
Yearly deposition
Risk atlas of potential threats from long-range atmospheric dispersion and deposition of radionuclides.
Sellafield nuclear fuel reprocessing plant
EnergyEnergy EnvironmentEnvironment HealthHealthCC
Hypothetical release of 100 g Anthrax spores
Monitoring stations
Source Determination by Inverse Modelling
Inhalation dose calculated by DERMA based on DMI-HIRLAM.
Determination of source location by adjoint DERMA using monitoring data. No a priori assumption about source (point, area, …).
EnergyEnergy EnvironmentEnvironment HealthHealthCC
Accidental fire in waste deposit Accidental fire in waste deposit.
Aalborg Portland, 23 October 2005 EnergyEnergy EnvironmentEnvironment HealthHealthCC
Accidental fire in waste deposit
DERMA calculations
Aalborg Portland, 23 October 2005
EnergyEnergy EnvironmentEnvironment HealthHealthCC
RoofWall
Street
Momentum Turbu-lence
Heat
DragWake diffu-sion
Radiation
Urban Features EnergyEnergy EnvironmentEnvironment HealthHealthCC
Urban Effects
The ABL height calculated from different DMI-HIRLAM data (left: urbanized, right: operational T). Main cities and their effect on the ABL height are shown by arrows.
EnergyEnergy EnvironmentEnvironment HealthHealthCC
Urban Effects
Local-scale RIMPUFF plume corresponding to a hypothetical release calculated by using DMI-HIRLAM data.
Cs-137 air concentration for different DMI-HIRLAM versions(left: urbanized 1.4-km resolution, mid: operational 5 km, right: operational 15 km).
EnergyEnergy EnvironmentEnvironment HealthHealthCC
City-Scale Obstacle-Resolved Modeling (TSU-CORM)
EnergyEnergy EnvironmentEnvironment HealthHealthCC
Streamlines and air pollution conc
3 d. fluid dynamic air pollution model
Resolution:Horizontal: 1x1 m2
Vertical: from 1m
Will be implemented spring 2007 at DMI and linked with DMI-HIRLAM, CAC and /or ENVIRO-HIRLAM.
DMIs Possible Modeling Activities In CEEH
EnergyEnergy EnvironmentEnvironment HealthHealthCC Kick-off.
d. 23-25/1 2007
Long-term simulations:• ENVIRO-HIRLAM and/or CAC.
Episodes:•ENVIRO-HIRLAM.
Modeling of the environmental impact of energy production/consumption
Long-term simulation of ENVIRO-
HIRLAM and/or CAC using HIRHAM
Meteorology.
Climate change impact on air pollutionand population health
DMIs Possible Modeling Activities In CEEH
EnergyEnergy EnvironmentEnvironment HealthHealthCC Kick-off.
d. 23-25/1 2007
Modify DERMA or CAC for sensitivi-
ty, risk/impact minimization and
optimization studies.Sensitivity studies for
environmen-tal risk/impact assessments.
Optimization modeling of environmentalrisk/impact studies
City scale modeling using TSU-
CORM.Link the air pollution
prediction from ENVIRO-HIRLAM or CAC to population activity (human
expo-sure modeling).
Human exposure modeling
The predicted exposure of population to NO2 (g/m3 *persons).
© Helsingin kaupunki, Kaupunginmittausosasto 576§/1997, ©Aineistot: Espoon, Helsingin, Kauniaisten ja Vantaan mittausosastot
Environmental OfficeKansanterveyslaitosFolkhälsoinstitutetNational Public Health Institute
EnergyEnergy EnvironmentEnvironment HealthHealthCC FUMAPEX integrated population health impact study